Role-Playing and Playing Roles: The Person,
Player, and Persona in Fantasy Role-Playing
Dennis Waskul
Minnesota State University, Mankato
Matt Lust
Southern Utah University
In fantasy role-playing games, participants collectively create and play fan-
tasy personas in an imaginary universe by using a vast system of rules that
function as guidelines for make-believe action and interaction. Conse-
quently, role-playing games obligate participants to occupy a liminal role
located in the boundaries of persona, player, and person. This study, based
on approximately ninety hours of participant observation and forty inter-
views with thirty role-players, explores how role-players actively negotiate
these symbolic boundaries: how role-players carve out distinct spheres of
meaning between themselves, their fantasy personas, and status as players
of these games. It also illustrates how these distinctions fail. Boundaries erupt
and role-players prove unable to compartmentalize themselves so discretely.
Through the lens of these games, we can examine simplified and exagger-
ated dynamics and entertain the possibility that we are all players located
at the liminal margins between the people we believe ourselves to be and
the personas we perform in situated social encounters.
Roles may not only be played but also played at, as when children, stage
actors, and other kinds of cutups mimic a role for the avowed purpose of
make-believe; here, surely, doing is not being. But this is easy to deal
with. A movie star who plays at being a doctor is not in the role of doctor
but in the role of actor; and this latter role, we are told, he is likely to take
quite seriously. The work of his role is to portray a doctor, but the work is
only incidental; his actual role is no more make-believe than that of a real
doctor merely better paid. . . . These desperate performers are caught ex-
actly between illusion and reality, and must lead one audience to accept
the role portrait as real, even while assuring another audience that the
actor in no way is convincing himself.
 Erving Goffman, Encounters
Direct all correspondence to Dennis Waskul, Minnesota State University, Mankato, Department of Sociology
and Corrections, Armstrong Hall 113, Mankato, MN 56001; e-mail: [e-mail].
Symbolic Interaction, Volume 27, Number 3, pages 333 356, ISSN 0195-6086; online ISSN 1533-8665.
© 2004 by the Society for the Study of Symbolic Interaction. All rights reserved.
Send requests for permission to reprint to: Rights and Permissions, University of California Press,
Journals Division, 2000 Center St., Ste. 303, Berkeley, CA 94704-1223.
334 Symbolic Interaction Volume 27, Number 3, 2004
This study examines roles, role-playing, and personhood in the context of popular
role-playing games. Role-playing games constitute a unique environment in which
fantasy, imagination, and reality intersect and oblige participants to occupy the role
of a  PC, gaming lingo for  player-character  a marginal hyphenated role that is
situated in the liminal boundaries of more than one frame of reality.  Games, as
Goffman (1961:27) wrote,  are world-building activities. Fine (1983:7) further sug-
gests:  By simplifying and exaggerating, games tell us about what is  real.  Taking
cues from Goffman and Fine, we seek to understand how participants in role-playing
games negotiate the precarious boundaries between reality, imagination, and fan-
tasy. We will conclude with commentary on what these simplified and exaggerated
 world-building activities reveal about how we all manage these kinds of distinc-
tions in more normative experiences of everyday life.
THE ROLE-PLAYING GAME: PERSONA, PLAYER, AND PERSON
FANTASY, IMAGINATION, AND REALITY
In the early 1970s Dave Arneson and E. Gary Gygax both members of Castles and
Crusades Society, an informal Minneapolis St. Paul gaming organization became
dissatisfied with the standard fare of medieval battle games. Sometime between
1970 and 1971 Arneson organized a unique game, the  Blackmoor Dungeon Cam-
paign, structured by principles we now deem fantasy role-play.1 After correspond-
ing with Gygax and additional play-testing, Dungeons & Dragons was first published
in 1974 by Gygax s company, TSR Hobbies, Inc. By 1979 Dungeons & Dragons was
selling seven thousand copies a month and declared by Fortune magazine the hot-
test game in the United States (Fine 1983; Smith 1980). In a few short years, Dun-
geons & Dragons bore a new genus of games and popularized an innovative method
of playing games.
Three decades later Dungeons & Dragons has been revised in new editions most
notably Advanced Dungeons & Dragons (post-1989) with expanded rules, elabo-
rated gaming environments, and enough exquisite detail to satisfy the most enthusi-
astic gamers. No other fantasy role-playing game has been as commercially success-
ful, and none has been quite as popular(ized). The commercial success of Dungeons
& Dragons has spawned competition; however,  D & D (as gamers affectionately
call it) remains a standard in what is now a crowded industry of fantasy role-playing
games. While the historical popularity of Dungeons & Dragons cannot be denied,
the purpose of our research is to understand the significance of the unique ways
these role-playing games configure fantasy, imagination, and reality as participants
necessarily negotiate between persona, player, and person.
Role-Playing and Playing Roles 335
The Persona, Player, and Person: Role-Playing and Fantasy Adventure
Fantasy gaming is a social world, luxurious in imagination and filled with
mysterious delights. This is a world of distant keeps, regal castles, glisten-
ing starships, fierce hippogriffs, rainbow dragons, and fiery jewels. It is
also a world of dank dungeons, villainous necromancers, green slime,
and omnipresent death. It is a world of dreams and nightmares; yet un-
like these constructions of our sleeping mind, these worlds are not experi-
enced in a state of reverie or unconsciousness. These worlds are experi-
enced collectively they are shared fantasies. This shared component
raises issues not present in private fantasies.
 Gary Alan Fine, Shared Fantasy
It is impossible to count the number of hours both authors have spent as dwarf,
elf, thief, magic user, or fighter. Over many years we developed untold numbers of
fantastic heroes who defeated legions of extraordinary beasts, pried wondrous trea-
sures from innumerable dead foes, and earned mammoth sums of gold as payment
for our explorations of perilous fantasy worlds where good and evil are in a con-
stant state of literal warfare. We delighted in these imaginative games, shared ad-
ventures with friends and acquaintances along the way, and, at least to some extent,
still indulge these delicious flights of fancy.
The first author spent much of his adolescent years playing Dungeons & Drag-
ons, continuing these games intermittently during his undergraduate college career.
After a decade-long hiatus, he joined a role-playing group as a participant observer
for the purposes of this research. The second author has been role-playing since
1998 and is involved in a broader range of contemporary games, including Mekton
Zeta, Dragon Quest, Big Eyes Small Mouth, Spy Craft, and Epic. We collected the vast
majority of data for this study from groups playing Dungeons & Dragons. Between
October 2002 and January 2003 we participated in approximately ninety hours of
fantasy role-play gaming sessions. We actively maintained field notes, but the pri-
mary source of data was forty open-ended qualitative interviews with thirty gamers.
Although games were sometimes held in private locations, the principal setting was
a local gaming store where we conducted fifteen- to twenty-five-minute interviews
before or after game sessions. Data were recorded by hand, verified for accuracy by
interviewees, and later analyzed for general patterns, trends, and themes.
Dungeons & Dragons is a dice-based role-playing game structured by guidelines
specified in  core rulebooks (as are all other games included in this study). Partici-
pants use dice to generate random numbers that correspond to the traits and abili-
ties of a fantasy persona. Once created, these fantasy personas are imaginatively
role-played. Yet the consequences and outcomes of make-believe role-play are al-
ways subject to indeterminate probabilities that are also mediated by the roll of dice.
Dice rolling maintains an element of tension and uncertainty, a key characteristic of
play (Huizinga 1950:47), assuring that  [t]here is always the question:  will it come
off?  Players use a variety of dice (four, six, eight, ten, twelve, and twenty-sided) and
336 Symbolic Interaction Volume 27, Number 3, 2004
core rulebooks specify which dice should be rolled in what conditions and how to in-
terpret the result of the roll. By using rules, dice rolls, and a hearty imagination play-
ers collectively generate  a habitable universe for those who can follow it, a plane of
being, a cast of characters with a seemingly unlimited number of different situations
and acts through which to realize their natures and destinies (Goffman 1974:5).
Dices are important in most role-playing games as they are the principal means
of simulating chance and probability. However, in the final analysis role-playing is
more like games of mimicry than either chance or competition (see Caillois [1958]
2001). Role-playing games are largely about fantasy: action occurs in make-believe
scenarios aptly described by Goffman (1974:46, 48) as an engrossable  realm. 2
These fantastic  realms of fantasy role-play are not only generated from rulebooks
and dice rolls but also by a  dungeon master,  referee, or  gamemaster. The game-
master occupies the most important role in fantasy role-playing games one that is
often described by players as  God-like. Gamemasters create the worlds, plots,
and scripts that generate a make-believe setting for game play. If player-characters
are told they are in a city located on an oasis in a vast desert, or that they encounter
a mysterious man who invites them to meet with the high priestess of the temple of
Venus, it is the gamemaster who not only creates these landscapes and situations
but also plays the role of the mysterious man, the priestess, and any other  non-
player-character that participants encounter. Similarly, if player-characters encoun-
ter hostile creatures, the gamemaster determines what kind of hostile creatures they
are, how they are armed, their combat, and any other actions they might take.3 In this
way, role-playing games are akin to improvisational theater: fantasy action collectively
sustains the dramatic narrative of a coauthored Goffmanian realm that is imagina-
tively fashioned by gamemasters and players through the use of dice and gaming rules.
Clearly, fantasy role-playing games are leisure activities that involve a unique
form of play.  The game is not competitive, has no time limits, is not scored, and
has no definitions of winning or losing. Unlike card games, board games, games of
chance, or organized sports, the point of fantasy role-playing games is neither
merely to play well nor to  win. Instead, the goals of the game are survival and
character development: participants create and play fantasy personas that, if kept
 alive, increase and advance skills and abilities over the course of many often-
lengthy gaming sessions. These personas fall into quasi-occupational classes (for ex-
ample, barbarians, assassins, or wizards) who have expertise in specialized skills and
abilities (such as spell-casting, pickpocketing, or the handling of medieval weap-
ons). Personas often belong to fantasy races humanoid beings that have their own
special  racial traits (such as elves, dwarfs, half-orcs, and halflings). Most impor-
tant, participants play fantasy personas: they bestow symbolic personas that are
fashioned in the liminal boundaries between interaction with other players during
the course of the game and fantasy action in a world of dragons, goblins, valiant
swordsmen, sagely wizards, and epic medieval warfare. Although the thematic set-
ting varies from one game system to the next, this liminal condition is generic to all
fantasy role-playing games and obligates participants to actively negotiate distinc-
tions between persona, player, and person.4
Role-Playing and Playing Roles 337
In role-playing games each participant is the fantasy persona he or she plays a
brutal barbarian, a mystical illusionist, a sly gnome.  For the game to work as an
aesthetic experience players must be willing to  bracket their  natural selves and
enact a fantasy self. They must lose themselves to the game (Fine 1983:4). Role-
playing games are  not  ordinary or  real life. It is rather a stepping out of  real life
into a temporary sphere of activity with a disposition all of its own (Huizinga
1950:8). However, since fantasy personas are played not merely generated by
rules and dice make-believe remains influenced by the same symbolic processes
that mediate nonfantasy public personas. In other words, role-playing games are
played with others who come to know fantasy personas (their own and others) on
the basis of a collective history of real and fictitious action and interaction.
A participant in role-playing games is also a player; the gamer who plays the imag-
inary persona. As a player, each participant must know and understand the rules of
the game that function as organizational guidelines for action and interaction. Players
must know which dice to roll in what situation, which rulebook to consult in what
circumstance, and how to manipulate a vast system of practical gaming knowledge
that specifies what a fantasy persona can and cannot do, when, where, and how.
Successful and satisfying games involve players who not only role-play but also pos-
sess proficiency in the complex rules. A participant in these games must not only
play the role of a fantasy persona, but the player as well.
Finally, and perhaps most ironically remote in these gaming sessions, each
 player-character is also a person. Participants in fantasy role-playing games are
not only personas and players; they may also be called students, employees, ado-
lescents, adults, spouses, parents, and a wide variety of other statuses they occupy
and roles they play in everyday life. As Fine (1983) has detailed, sometimes these
other self-investments can interfere with role-playing games and vice versa. How-
ever, for the most part, role-playing games are fantasy adventures (Simmel [1911]
1971) or activity enclaves (Cohen and Taylor 1992). They are hobbies a form of
recreational leisure a distinct sphere of activity that is segregated from the nor-
mal strictures of life; activities most people engage when not preoccupied with
routine involvements that otherwise describe mundane life. Consequently, these
kinds of activities are  outside and above the necessities and seriousness of every-
day life (Huizinga 1950:26). Like most hobbies or leisure activities, fantasy role-
playing  is essentially a separate occupation, carefully isolated from the rest of life,
and generally is engaged in with precise limits of time and place (Caillois [1958]
2001:6).
Role-playing games can be described, explained, and understood as an activity
that exists in the unique interstices between persona, player, and person. How do
participants in fantasy role-playing games negotiate these liminal symbolic bound-
aries? To what extent do these decidedly playful negotiations illuminate the ways
we all actively fashion the precarious distinctions between person and public per-
sona? Since all people necessarily juggle a multiplicity of roles sometimes shifting
from one to the next with remarkable fluidity are not we all players of fantasy
role-playing games?
338 Symbolic Interaction Volume 27, Number 3, 2004
Roles, Play, and Role-Playing: Reality, Imagination, and Fantasy
The mental sphere from which the drama springs
knows no distinction between play and seriousness.
 Johan Huizinga, Homo Ludens
All people play and play with roles: we take up, define, and negotiate a wide
array of social roles that, though often structured in meaningful and consistent
ways, are enacted uniquely from one person to the next. People play roles, and roles
play a significant part in defining self. Just as we actively and fluidly construct the
roles we play, those roles also define and structure self in broad social, cultural, and
temporal frameworks of meaning. Of course, George Herbert Mead (1934) clearly
perceived the dual and pivotal character of social roles; they are central to his
understanding of the fundamental relationships among mind, self, and society.
Mead also understood the significance of play and games to both acquiring a self
and developing the capacities for selfhood. Self is accomplished in a process of
 taking the role of the other with increasing degrees of sophistication that are mas-
tered in sometimes literal and other times metaphorical play and gaming activities
(Mead 1934). Children literally play imaginary roles and in the process develop the
capacity to see themselves as others might to see themselves as both object and
subject the quintessential quality of self. While often less obvious and less literal,
the same dynamics occur in experiences that extend well beyond childhood. An adult
acquires a new self through a process akin to the play and games of children: imagi-
nary roles are evoked, and these sentiments (Cooley [1902] 1964) provide a structure
of meaning for playing at a self that has not yet merged with the person (Turner
1978) but will, over time, be mastered in increasing increments of sophistication.
From this perspective, play and games are distinct forms of role-playing activity
that present a distilled lens for better understanding the relationships among fan-
tasy, imagination, and reality. This is partly what Mead implied in his discussion of
play and game5 and certainly what Goffman meant when he wrote:
It is only around a small table that one can show coolness in poker or the capac-
ity to be bluffed out of a pair of aces; but, similarly, it is only on the road that the
roles of motorist and pedestrian take on full meaning, and it is only among per-
sons avowedly joined in a state of talk that we can learn something of the mean-
ing of half-concealed inattentiveness or relative frequency of times each individ-
ual talks. (1961:27)
In this way, the presumably distinct categories of fantasy, imagination, and reality
can be shown as a subtle continuum of finely graded experience. More precisely, all
social reality can be understood as emergent from the interstices of these inter-
related provinces of meaning.  Conceptions are thus born as acts of the imagination
(Huizinga 1950:136).
The interrelatedness of fantasy, imagination, and reality prove central to interac-
tionist articulations of social reality. A fundamental tenet of symbolic interaction is
that human beings do not experience reality directly but through symbols, language,
Role-Playing and Playing Roles 339
social structure, and situated variables of social interaction. Mead s  rejection of the
realists position which asserts that what exists inside and outside the mind is iso-
morphic (Maines, Sugrue, and Katovich 1983:164) is one of the most important or-
ganizing themes of symbolic interaction:  there is a world which subsists, but does
not necessarily exist (Mead 1936:336). Thus  the world of illusion should be in-
cluded in the structure of society. People create illusions and then induce others to
impute meaning to them and act in accordance with those meanings (Maines, Sug-
rue, and Katovich 1983:170). Consequently, fantasy, imagination, and reality are no-
toriously porous: experience, knowledge, and understanding routinely slip from one
to another.
In the lived experience of everyday life just as in play and games fantasy, imag-
ination, and reality are not so easily compartmentalized but necessarily blend and
blur to such an extent they are often difficult to convincingly separate into mutually
distinct categories. Contemporary interactionist literature is filled with examples of
this blending and blurring. For example, borrowing significantly from Cooley s
looking-glass self ([1902] 1964), Hertz (2002) details how families created by anony-
mous sperm donors actively construct imaginary fathers from the most minuscule
scraps of information. Mothers, and eventually children, craft stories about these
 ghost fathers, and, in the process,  the anonymous donor takes on a persona of
his own a person who may be more fiction than fact (Hertz 2002:6). Built on
solid interactionist foundations, Hertz s analysis hinges on a powerful insight:
fatherhood is an idea that exists independent of a father, and the idea of fatherhood
is just as important, if not more so, than fathers themselves. The absence of an
 actual father makes the looking-glass of fatherhood all the more apparent.
Hertz concludes her analysis by identifying how and why these ghost fathers
affirm and inform important dynamics of contemporary postmodern families. There
is, however, a much more provocative and much less  post implication:  actual
copresent fathers may be just as ghostly as the fathers of children conceived by
anonymous sperm donors. In all cases, therefore, fatherhood is defined in a process
that includes the fictions of looking-glass idealism. Given Hertz s analysis, it is easy
to see how the symbolic role of father is not only distinct from men themselves but
perhaps more important to the processes of pinning down a self. In noting the same
dynamic for motherhood, Carse suggests this characteristic defines all social roles:
It is in the nature of acting, Shaw said, that we are not to see this woman as
Ophelia, but Ophelia as this woman. . . . To some extent the actress does not see
herself performing but feels her performed emotion and actually says her mem-
orized lines and yet the very fact that they are performed means that the words
and feelings belong to the role and not to the actress. . . . So it is with all roles.
Only freely can one step into the role of mother. Persons who assume this role,
however, must suspend their freedom with a proper seriousness in order to act
as the role requires. A mother s words, actions, and feelings belong to the role
and not to the person although some persons may veil themselves so assidu-
ously that they make their performance believable even to themselves, over-
looking any distinction between a mother s feelings and their own. (1986:15 16)
340 Symbolic Interaction Volume 27, Number 3, 2004
The conclusion is clear: to some extent we are all participants in fantasy role-
playing games. Father, mother, professor, student, sociologist even symbolic inter-
actionist all words to fashion symbolic self-claims in reference to social roles and
statuses; uniquely situated provinces of meaning that, as Hertz (2002:3) described,
are often  more ghost like than real. We must be cautious and not push this conclusion
too far. As Goffman (1974:2) wisely wrote,  social life is dubious enough and ludi-
crous enough without having to wish it further into unreality. We merely suggest
that paramount reality is not distinct from fantasy and imagination, which are among
its most interesting and fluid dimensions.6
In role-playing games participants are uniquely situated in the loose boundaries
of the person-player-persona trinity. It should be clear by now that the distinctions
and permeable boundaries between person, player, and persona roughly adhere to
the more general trinity of reality, imagination, and fantasy. Participants in fantasy
role-playing games literally and consciously play with this trinity of social reality;
the significance of this research is that those same porous distinctions and active ne-
gotiations also occur in everyday life. Thus the question that guides this research is
both simple and complex: how do people negotiate these explicitly playful and de-
cidedly fantasy games? Clearly, answers to this question may implicate normative
experiences of everyday life that are often more serious, sometimes more real, and
occasionally less ludic.
PRECARIOUS BOUNDARIES AND FANTASY ROLE-PLAY
Because fantasy role-playing is structured by the rules of a complex game, we begin
our analysis by discussing the general nature of these games. Our intent is to provide
enough detail to clarify game-play while also identifying unique characteristics salient
to how gamers create and play fantasy personas. The balance of our analysis focuses
on how participants in role-playing games negotiate person-player-persona sym-
bolic boundaries and the extent role-players are able to maintain these distinctions.
Fantasy Role-Playing Games and Gamesmanship
In role-playing games, players use a complex system of rules to craft fantasy per-
sonas in a fantastic universe of make-believe.7 In practice, however, these rules are
less regulatory and more a set of conventions and guidelines that provide a struc-
ture for exquisite detail. In other words, players use  rules as gaming resources
rather than gaming limitations, and most experienced role-players understand that
 there are no rules that require us to obey the rules (Carse 1986:10).  One of the
cardinal  metarules of FRP [fantasy role-playing] gaming is that there are no
 rules ; the rulebooks are only guidelines (Fine 1983:115). Rather than being bound
by rules, role-playing games are structured by conventions that loosely define basic
persona traits and qualities of a make-believe world that participants play at and
game with which is exactly what David claimed:
Role-Playing and Playing Roles 341
Role-playing is enjoyable because I m no longer bound by the rules so to speak.
In role-playing games I can be a wizard, a fighter, a cleric while being a dwarf,
human, elf, or half-orc. Because we don t live in the world that we role-play we
are able to bend the rules to fit how we want to play. But, in reality, you can t
bend the rules. You can t hover, or throw fireballs, or take a hit from a giant or
an ogre but in role-playing games you can. That s what makes them fun.
Participants create fantasy personas from basic attributes generated by random
dice rolls that players interpret by assigning their personas varying levels of strength,
intelligence, wisdom, dexterity, constitution, and charisma allowing them to create
imaginary personal characteristics that are best suited for the kind of fantasy per-
sona they would like to play. A player who intends to develop a wizard or illusionist
needs a persona with great intelligence and wisdom, thieves and assassins need dex-
terity and charisma. As James remarked:
I usually construct characters according to how I hope to play them. If I think
I m going to play a strong fighter I m going to give him a lot of strength and con-
stitution points but not a lot of intelligence or wisdom. These aren t important to
a fighter.
Once created, participants role-play the words and actions of their persona. Akin
to discursive impromptu acting, for the most part role-playing is unlimited, con-
strained only by an unspecified yet shared sense of naive realism. Fantasy personas
may say and do whatever they please, so long as other players and the gamemaster
agree that such actions are  reasonable. For example, if a player-character is told
that he8 notices a bright shiny ring at the bottom of a pool of water, he merely needs
to announce that he will dive into the pool, swim to the bottom, and retrieve the
ring. The same player-character could not walk on imaginary water (without magi-
cal aid, which is possible in these games) but can go for a swim at any time. The lat-
ter action is perfectly  reasonable ; it adheres to a basic sense of naive realism and
is thus considered appropriate role-playing.
However, the game becomes much more complicated. Like real life, actions have
consequences, and most  significant actions (such as combat moves, spell casting,
the use of specialized skills) depend on conditions that do not always guarantee suc-
cess. Dice rolls largely determine these variable outcomes and consequences. For
example, the player-character who dove into a pool to retrieve a glimmering ring
may have failed to announce that he will remove his armor before jumping into the
pool an oversight that could have serious consequences. Although none of us
knows with absolute certainty, naive realism suggests that it is difficult to swim while
suited in battle armor. Thus the gamemaster will instruct the player to roll dice to
determine if the fantasy persona will recover from his blunder or sink to the bottom
of the pool. Even if the player remembers to remove his armor, he may swim to the
bottom, grab the ring, and suddenly discover that it is a decoy placed by some mis-
chievous agent of evil: the ring is a trap that has been unwittingly sprung by his
touch. Once again, the player will be instructed to roll dice to determine if he is able
to escape the trap or will be ensnared in a watery grave. These kinds of circum-
342 Symbolic Interaction Volume 27, Number 3, 2004
stances, their parameters, the rules for rolling which dice in what situation, and how
to interpret the outcome are detailed in core rulebooks.
For all practical purposes, this is how the game is played. Players describe what
their fantasy personas say and do. In the case of significant actions the roll of dice
determines the outcome, which then compels further actions contingent on the results
of an ongoing chain of imaginary action, outcome, and reaction. Combat situations
which comprise a large part of these games work in much the same way. For ex-
ample, if a player-character encounters an unarmored drunk at a local tavern who
proceeds to berate him with an unrelenting stream of insults, the player may decide
that his fantasy persona s honor has been offended and announce he will stand up,
walk over to the drunk, and slap him across his irreverent face. As in real life, we
can choose these kinds of actions any time, but actions and intentions are not one
and the same. The fantasy persona may intend to slap the drunk across his face, but
his actions are merely an attempt; his slap may hit or miss. Thus the player rolls dice
to determine if his fantasy slap  hits. Because the man is drunk and unarmored, he
is an easy target; the player may, for example, need to roll a five or higher on a
twenty-sided dice (an 80 percent chance) to guarantee success. If the same fantasy
persona attempted to slap a palace guard who is well protected, trained, and pre-
pared to deal with these kinds of shenanigans the odds of  hitting substantially
decrease; the player may need to roll eighteen or higher on a twenty-sided dice (a 10
percent chance of success). If a player-character successfully  hits, another dice
roll determines how much  damage is delivered. A slap produces little damage
(say, for example, the roll of a four-sided dice), a short sword will do more (the roll
of a six-sided dice), and a dwarven waraxe is even deadlier (the roll of a ten-sided
dice). These imaginary people (or beasts) may attack in turn, and opponents roll
dice (actually rolled by the gamemaster) that determine if they  hit and, if so, how
much  damage results. The number needed to  hit is determined by many factors,
including the kind of armor skills and level of the opponent, and  damage is deter-
mined by the type of weapon used modified by the strength of the attacker.
Fantasy personas and opponents have a certain number of  hit points (also gen-
erated by random dice rolls that accumulate as the character advances). The more
 hit points a persona accumulates, the more  damage he or she can sustain. In
combat, damage is subtracted from hit points; and when hit points reach zero he,
she, or it falls  unconscious and is declared  dead at negative ten. The basic ob-
jective of these games is to keep fantasy personas alive through numerous encoun-
ters like these, and doing so requires intelligence, skill, knowledge of gaming rules,
and creative problem solving. At the very least, players must become adept at glean-
ing the right clues in order to reasonably size up potential opponents; a level one
player-character will be utterly destroyed by a level ten opponent. A key to success-
ful gaming is learning how to make these critical judgments as player-characters
navigate the dangerous and typically violent worlds of fantasy adventure.
Our description of game-play is woefully simplified and does considerable injus-
tice to the actual complexities of the game. A more likely scenario is one where the
Role-Playing and Playing Roles 343
unarmored drunk at the tavern turns out to be a high-level mage who, in spite of his
drunkenness and foul mouth, is neither to be trifled with nor is he inclined to kindly
turn his cheek. When a player-character becomes ensnared in a trap, he is rarely
declared instantly  dead ; these circumstances only cause the other player-characters
to spring to action and try to rescue their unwittingly helpless companion. In other
words,  the game is an ongoing coauthored narrative that players fashion out of the
enormous possibilities for dramatic imaginary actions, consequences, and reactions
that are mediated by probabilities determined by the roll of dice. That is what the
game is all about: teamwork, cooperation, and survival are the organizing themes.
Although simplified, this brief description adequately highlights two critical
characteristics of game-play. The first is that role-play is in the form of discursive
impromptu acting: players describe what their fantasy personas say and do in the
various situations they encounter and how they respond to the myriad ongoing con-
sequences that result from those actions. Second, game-play involves rules and
guidelines for dealing with chance, probability, and random outcomes all of which
are mediated by the roll of dice. In this way, infinite possibilities for imagined action
intersect with finite yet indeterminate probabilities and random chance. Finite guide-
lines generate a structure for infinite play,  an open-ended game that any number
can play forever (Goffman 1974:6). Thus no two players can play the same fantasy
persona in identical fashion, nor is it possible for two identical situations to result in
precisely the same outcome. For this reason, the game is exceptionally  life-like 9
 more like life, and less like games (Fine 1983:8). As Trent and Justin told us:
I enjoy rolling the dice because I like the fact that I can t control everything. . . .
Chance is so important because it is the only way to really simulate reality in the
game setting. I mean life doesn t really happen according to how we really want
it to, so chance helps to keep things pretty real.
Just like in life quite often the unexpected can really change things that you
never expected to change, and change them in ways that you never could have
expected. This is what makes role-playing such a really wonderful time you
never know what s going to happen next.
All a player can know are the rules of the game, which detail probabilities for the
various actions characters might perform. The development of a fantasy persona
depends on how the player handles the outcomes of these probabilities, which al-
ways entail uncertainty and chance. The realm of fantasy role-play much like a
Schutzian Lebenswelt or  life world   is something that we have to modify by our
actions or that modifies our actions (Schutz 1973:209). Consequently, the fantasy
personas of role-playing games are not unlike people in everyday life chiefly in-
fluenced not by the basic traits they start out with but by the choices they make, the
outcome of those decisions, chance, and the ongoing dialectical relationship be-
tween consequences and personal adjustments. This fluid dynamic is precisely what
Steve and Mark indicate:
When you roll a character its just paper, but what happens shapes what the char-
acter ends up being.
344 Symbolic Interaction Volume 27, Number 3, 2004
I usually just play the character how he s rolled and after a few hours I ll start
evolving according to how he s been going. I adjust to how he s been reacting to
others and how others have been adjusting to him.
Like everyday life, fantasy personas in role-playing games emerge from the innu-
merable possibilities that culminate over a history of choices, decisions, and conse-
quences that are patterned and structured, yet also unavoidably unpredictable and
indeterminate. Mike made this point:
Simply because you roll up a character and put his or her stats down on paper
doesn t mean he or she has any kind of personality yet. You have to play the
character in order to develop a character s real personality. (Emphasis added)
The Persona, Player, and Person: Negotiating Borders and Boundaries
[A good role-player is] someone who plays in character and doesn t let
player knowledge interfere with character knowledge doesn t let what
happens in the game interfere with playing the game on a player level.
He plays the game in the game and doesn t bring personal problems into
the game. It s no fun when someone does that because it plays the game
out of characteristics.
 Dan, Dungeons & Dragons player
Because role-playing games are situated in more than one frame of reality, the
activity involves more role-play than most participants recognize. While role-players
tend to think of role-playing as something restricted to the moments when they play
a fantasy persona, it is clear that the general dynamics of role-playing involve much
more. In fantasy role-playing games, participants must actively establish symbolic
boundaries between player, persona, and person and assume the right role in the
right condition a circumstance that evokes border-work.
While the concept of border-work is most often used to examine the dynamics of
interpersonal relations (tactics that establish and maintain distinctions and bound-
aries between people), in many circumstances it also involves important intraper-
sonal boundaries. By  intrapersonal boundaries, we are loosely referring to  the
organization of experience something that an individual actor can take into his
mind and not the organization of society (Goffman 1974:13). However, distinc-
tions between inter- and intrapersonal border-work are purely conceptual. In prac-
tice, inter- and intrapersonal border-work are quite permeable. When interpersonal
border-work demarcates symbolic territories of truly different situations, these bound-
aries may also evoke intrapersonal forms of border-work that are necessary for the
truly different selves required of those situations. This is precisely the situation of
fantasy role-playing games: inter- and intrapersonal boundary work become neces-
sary because while some circumstances require participants to be in persona, others
require a player who must control the non-game-related aspects of his person to
prevent them from interfering with game-play.
Role-Playing and Playing Roles 345
Of the three, the non-game-related aspects of the person are the most potentially
disruptive. This is not surprising, since these are role-playing games. By design and
intent, participants are expected to be players or the personas they are playing. Or,
as William told us,  You re not yourself, you re playing someone new. . . . That s the
whole thing about role-playing; you re not there to play yourself, you re there to
play someone else. It is understandable why the nongame aspects of the person are
irrelevant, distracting, and extraneous. James echoed this idea when he said that a
good role-player is  someone who doesn t let personal feelings interfere with the
game. I leave work at work and home at home same principle in gaming. Indeed,
many participants in this study used the same analogy of  leaving work at work as
part of their definition of a  good role-player or a  good gaming session. Thus
participants in role-playing games may fluidly move between player and persona,
but other aspects of personhood are more carefully contained.
For the most part, players bracket their persons with relative ease; it is implicit in
the social structure of the gaming sessions themselves. As we wrote in the field
notes from one of our first gaming sessions, role-players are often a motley crew of
dissimilar people who are otherwise separated by significant social, cultural, and
institutional barriers:
It is an unlikely mix of people who have somehow come together to play this
game . . . a university professor, a few university students, a few high school stu-
dents, and others whom I could not place. Ages seem to range somewhere be-
tween an approximate sixteen to early thirties. The Dungeon Master shows up
still wearing his McDonald s work uniform, a couple players are wearing unre-
markable T-shirts and jeans, one player sports a derby hat and long black trench
coat, while another wears shaggy hair (partially colored and partially braided),
overly baggy clothes and hemp jewelry. . . . On the surface they appear to have
nothing in common, aside from the fact that they all carry Dungeons & Dragons
paraphernalia (books, gaming dice, character sheets, and miniature figurines).
By all indications, the only commonality among these people is their interest in
Dungeons & Dragons. Yet, instead of hindering social interaction and group forma-
tion, these differences proved instrumental even crucial if for no other reason
than players come to know one another in the course of game-play, leaving little
else to otherwise bind them in what becomes an unambiguously utilitarian relation-
ship. As we further noted:
At no point did anyone discuss issues of relevance to their work, family, school,
or anything else that pertains to their life outside of this game. Indeed, in spite of
the fact that there were new players present, no introductions were made, real
names were not shared, and nothing was mentioned about players as people. . . .
Since informal  get to know you chitchat seems to be either unimportant or ir-
relevant, I decided not to ask. But even more, normal conversation based on in-
teractive cues seems strangely uncouth. It does not seem appropriate to actually
ask the Dungeon Master if he does, in fact, work at McDonald s. It doesn t even
seem appropriate to introduce myself to these players, nor does it seem unusual
that they have not introduced themselves to me. Instead, players introduce them-
selves as the character they play during the course of gaming. I only know these
346 Symbolic Interaction Volume 27, Number 3, 2004
people by the character they are playing and they only know me as Cantrall a
rather standoffish fighter who, although brutish in appearance, reliable in com-
bat, and generally cooperative, does not get involved in the  party politics. 10
Not only do players come to know each other during the course of game-play,
but also, like the  friendly poker game (Zurcher 1983:138), during role-playing
sessions,  it was understood that there were to be no  outside interruptions. There
were no radios or televisions playing, no wives serving beverages, no children look-
ing over shoulders. In fact, for one of the role-playing groups included in this study,
it was necessary to repeatedly move the location of gaming sessions for no other
reason than the struggle to find a setting free of these distractions a context insu-
lated from  outside interruptions that not only interfere with gaming but might
also evoke roles superfluous to the game. As Huizinga (1950:21) noted,  The play-
mood is labile in its very nature. At any moment  ordinary life may reassert its
rights . . . which interrupts the game . . . by a collapse of the play spirit, a sobering, a
disenchantment (original emphasis).
To guard against these potential interruptions, role-playing sessions are ephemeral
situations encased not only by a  spatial separation from ordinary life (Huizinga
1950:19) but also by symbolic boundaries that  declare as irrelevant [the] norms and
roles that society at large deems mandatory in favor of idiosyncratic group norms and
roles (Zurcher 1983:154). However, in role-playing games the bracketed irrelevance
of the person is much more exaggerated than what Zurcher observed in the friendly
poker game. While it is, for example, rare for new players to introduce themselves to
others, it is common for participants to come to know each other only as the fantasy
persona they play a dynamic also noted in Fine s study of fantasy role-playing games:
As a new player I was struck by how little I learned about the private lives of
others even others to whom I felt close. One didn t talk about occupations,
marital status, residence, or ethnic heritage. In some cases it was months before I
learned a player s surname. Others confirmed this observation, and suggested
that it represented a need to establish a distance from one s real self. (1983:55)
This bracketing of personhood fosters a kind of  focused gathering that Goffman
(1961b:17 18) describes as providing a  heightened and mutual relevance of acts; an
eye to eye ecological huddle that is conducive to the experience of a gratifying  we
rationale. Or, as Schutz ([1932] 1967:164) might describe it, this kind of focused
gathering represents umwelt built of a pure yet also ephemeral we-relationship  in
which the partners are aware of each other and sympathetically participate in each
other s lives for however short a time. As these role-players illustrate, an essential
component of these focused gatherings is  rules of irrelevance (Goffman 1961b:26);
a  set of rules which tells us what should not be given relevance while also clearly
identifying  what we are to treat as real. On the basis of these implicit rules, role-
playing games occur within an  interaction membrane (Goffman 1961b:65) that
like friendly poker games  strengthen idiosyncratic norms and the cohesion and
 separateness by declaring irrelevant certain characteristics of the participants or set-
ting that may have considerable saliency in the world  outside  (Zurcher 1983:148).
Role-Playing and Playing Roles 347
The bracketing of person from both player and persona is implicit in the activity
itself. Although these dynamics have surprised some sociologists of everyday life
(see Fine 1983:55), they are not unusual. Whether playing a game (such as poker or
Dungeons & Dragons) or more serious roles within institutions or occupations, part
of what is implied in playing a role is that we are not playing others. The fact that
non-game-related aspects of a person are effectively bracketed or otherwise ignored
in role-playing games should not be any more surprising than the fact that marital
roles are often suspended when people are at work, work roles are often suspended
when people are at home, and so on. It is conventional to routinely encounter people
whom we know only as occupants of certain statuses, and these kinds of encounters
are quite normative. As the participants in this study told us repeatedly, the same
principle  leave work at work  applies equally to role-playing games.
Even so, role-playing games become much more complicated because this activity
necessarily involves a participant who actively occupies two distinct simultaneous
roles within the same activity; he is the fantasy persona he plays and the player who
enacts the persona.  There are two performances occurring in a role-playing game:
a collectively imagined theater of characters and events shared among the players
and the gamemaster, and the set of actual audio visual event that transpire among
the players and the gamemaster (Mackay 2001:89 90). Although this fine distinction
may seem purely academic, in practice the difference proves salient among fantasy
role-playing gamers. As James reported:  I try to separate myself from my character.
When something happens . . . instantly you as a player will react. [But] you [the player]
need to be careful how you [the character] react and distinguish between the two.
Note the words James uses to describe the distinction between player and per-
sona, and also that it is necessary to clarify what James means by adding more
specific information in brackets. In everyday life words like you and me are sufficiently
precise indicators of self. When  you ask  me a question it is clear who is inquir-
ing of whom, and it would be unusual for  me to wonder which  you is asking the
question or which  me ought to respond. However, in fantasy role-playing these
words can be ambiguous in a peculiar way. Because participants are simultaneously
both players and the fantasy personas they play, there exists a multiplicity of  you s
and  me s. It is not always clear which  you or what  me is being evoked. Even the
authors of the Dungeons & Dragons Players Handbook (Cook, Tweet, and Williams
2000:6) recognize this ambiguity and seek to distinguish player from persona:
The action of a Dungeons & Dragons game takes place in the imaginations of
players. Like actors in a movie, players sometimes speak as if they were their
characters or as if their fellow players were their characters. These rules even
adopt that casual approach, using  you to refer to and to mean  your character.
In reality, however, you are no more your character than you are the king when
you play chess. Likewise, the world implied by these rules is an imaginary one.
This precarious distinction between player and persona is crucial to role-playing
games.  The character identity is separate from the player identity. In this, fantasy
gaming is distinct from other games (Fine 1983:186; original emphasis). As Dave
348 Symbolic Interaction Volume 27, Number 3, 2004
stated,  I separate myself from my character. Some of the things I may consider log-
ical my character may not. Sometimes they do coincide. But I can t play me and my
character [at the same time]. Peter adds:  I try to think within the game as much
[like] my character as I can because there are certain things [that], as a player, I
wouldn t do but my character would do. I have to be careful in distinguishing
between the two or the game probably wouldn t be fun.
While on the surface Peter s statement appears to reiterate the ways that role-
players bracket non-game-related aspects of personhood during the course of game-
play, the situation is knottier: role-players must distinguish between the knowledge
they have as a player and the knowledge they presume their fantasy persona has.
Distinguishing between  player-knowledge and  persona-knowledge is necessary in
order to, as Chris said,  play their character as their character. In fact, this especially
perceptive role-player went so far as to define this quality as  metagaming. Chris
defines a lack of metagaming as circumstances where  you use player-knowledge
instead of character-knowledge, thus resulting in  bad role-playing . . . that will
ruin the game. He provides an example:
A player may know the hit points of an ogre because you as a player just read
the Monster s Manual and are transferring that knowledge to your character. . . .
[T]he Dungeon Master plans a game based on what the characters know. So it
can ruin the game. If a character doesn t know a monster has invulnerability
against fire he might just bring fire-based weapons instead of something else. But
if he does know that strength he will prepare against it.
This poses a dilemma for role-players:  It is a difficult moral decision for a player
not to use a solution to a problem because his character would not have thought of
it (Fine 1983:211). Yet the participants in this study consistently cited this moral di-
lemma as the key to good role-play. In the words of one role-player, Rodney, when
players do not separate player-knowledge from persona-knowledge when they do
not  metagame appropriately  it turns the game into dice rolling instead of role-
playing. Isaac described a situation in which failure to segregate player- from persona-
knowledge spoiled an otherwise good time:
Once when we were fighting an army of goblins well maybe an army is overex-
aggerated, but anyway because one of the players knew the average hit points
of a goblin and knew the average damage of his fireball spell, he knew exactly
how many times he would have to cast the spell. While it could be seen that the
player would know this, it seems that the player took the role-playing out and
turned it into a numbers game which, in my opinion, takes the fun out of the
game!
Rodney adds:  [A] minimum/maximum penchant can leave the character as a
statistic rather than a character. I like to embrace the class within the system and try
to find an aspect of that class I want to focus on and develop my character out of
that. Charlie summed it up neatly:  Role-playing is by definition playing the role
of another person. To play that person you have to keep their knowledge, values,
and motivations in mind, and react accordingly.
Role-Playing and Playing Roles 349
Porous Borders and Erupting Boundaries
I think that player-knowledge and character-knowledge should be kept
very separate, but it s impossible no matter how hard I try. And, it s really
important that I keep them separate because a game can be ruined by too
much player-knowledge seeping in.
 Trent, Dungeons & Dragons player
Although the analogy used in the Dungeons & Dragons Players Handbook, that
 you are no more your character than you are the king when you play chess, holds
conceptually, in practice it fails wretchedly. Role-playing games are not board
games; they differ in important ways that are best described by Fine:
In board games, such as backgammon, dice determine the outcome of sequential
action, but in fantasy games, unlike in backgammon, the dice generate actions
that could occur in the real world. A roll of six in backgammon means that the
player s piece gets to advance six spaces on the board; that same six in fantasy
gaming means that a player s character successfully bashes an opponent. While
both of these actions are unreal, they are unreal in different ways. In backgam-
mon, the pieces do move six spaces a physical movement of a material object
but the spaces have no inherent meaning. No physical movement occurs in fan-
tasy gaming, since the actions of characters are internally represented; however,
within the framework of the game the bash is a real one, and the character who
is bashed is really injured. The world of fantasy gaming and the rules that struc-
ture that world do not have physical effects, but the consequences are close sim-
ulations of natural interaction. The action is a direct stimulation of a hypotheti-
cal world rather than, as in backgammon, an indirect simulation enacted in a
physical world. (1983:184; original emphasis)
Because role-playing games necessarily involve impromptu discursive acting in
circumstances that are mediated by rules of probability and chance, they create a
unique set of social-psychological conditions that further distinguish them from
games like chess or backgammon. Although the game is purely fantasy, players must
act, interact, and react by imagining how they would handle the same circumstances
if they were their fantasy persona and the situations they encounter were genuine
by definition, that is what is implied by a role-playing game. As William stated,  I try
to be the character as much as possible. But in absence of a reference point for the
character s thoughts, it d be my own thoughts and reactions that come into play.
Dan reiterated this point:  As I play the character I think what I would do in this
situation. In this way, the neat distinctions between person, player, and persona
erode into utterly permeable and interlocking moments of experience. As Goffman
(1974:47; original emphasis) notes,  Fanciful words can speak about make-believe
places, but these words can only be spoken in the real world. Or, as Fine (1983:183)
explains,  by playing fantasy games, participants implicitly agree to  bracket the
world outside the game. Yet ultimately all events are grounded in the physical
world. Thus rigid distinctions between fantasy, imagination, and reality between
person, player, and persona prove untenable. Instead, role-playing games neces-
350 Symbolic Interaction Volume 27, Number 3, 2004
sarily involve, to borrow from Mead (1934),  taking the role of the other. In this
case, however, the  other is not someone else at all; the  other is a fantasy charac-
ter who is in fact the player and person himself. As several players told us, this pecu-
liar dynamic does not escape the attention of role-playing gamers:
You can t say that your PC [player-character] will never be an extension of your-
self because you are playing your character. If you think your character is sup-
posed to act that way or this way it s still just your perception because no one
else can take that same character and play it totally the same. . . . You can never
think like the character because the character is you. Whatever the character
thinks is coming from you so it is inherently a mixture. I could play a direct op-
posite from me for example, a female evil priestess and play it well in character
and still the actions would be coming from me. . . . No matter what you do it is
tied with you. It s kind of difficult to separate sometimes.
The reactions that my character takes I think or agree with, but the actions and
actual role-playing are my character.
In terms of character development I ve never really regulated it. Some Dungeon
Masters require one or two pages of character history. [However,] between cre-
ation and personality development, I try to find my characters in the playing of
them and I try to think to myself when making a decision, So if I do this now, is this
what I will always do, sometimes do, never do again? How will the actions I am
about to take [apply to my character], does it fit with what I ve done in the past?
Even more, participants in role-playing games often find it difficult to play a fan-
tasy persona that is purely fictional. Justin told us,  I m not very good at making
characters radically different than me. . . . Role-playing games are a fantasy projec-
tion of myself me having adventures I wouldn t normally have. In fact, many
role-players claimed that effective play presumes gamers who identify with and other-
wise apprehend the fantasy persona as an extension of themselves.  If a player
doesn t care about his character then the game is meaningless (Fine 1983:185).
Peter not only told us,  I find it funner to play characters I understand, but also
went on to describe other consequences of playing a persona that he does not iden-
tify with. He illustrated this idea by telling about a session when he played a paladin
but found it difficult because  law and order are beyond me to understand. Fur-
ther describing the situation, Peter said,  When we started he wasn t like me, but as
time went on he became more and more like me. He began as a defender of justice
but ended up a guy with a guilt complex. In short, he told us about how he cre-
ated a do-gooder fantasy persona but found it difficult to actually defend good-
ness and justice in the course of the game. Because he ended up role-playing in
ways that were out of moral alignment for his persona  he felt guilty. The irony is
that fantasy personas are purely fictional and thus cannot  feel any more guilt
than the player who plays them. Does the persona have a  guilt complex, or is
Peter merely guilty about how he has played him? Clearly, the answer is an am-
biguous both but neither. His persona has a  guilt complex and the player feels
guilty about how he has played him the guilt is real and exists in two simulta-
neous frames of reality.
Role-Playing and Playing Roles 351
As this evocative situation illustrates, participants in role-playing games are lo-
cated at intersections between person, player, and persona in a manner that funda-
mentally blurs the distinctions between them. Isaac gave us an additional anecdote
that illustrates the sometimes curious ironies of role-playing:
In Charlie s Spycraft campaign I play a face man character which requires that I
develop multiple personas to use at various times. What I usually do is choose a
name like Jeremiah Bell, for example. Then I create a personality, well, more of
a persona for this name from asking myself as a player but also as a character, Is
he rich? Is he poor? Is he smart? Is he backwoods? These are all questions that I
as a player and as a character have to answer to develop my character fully.
In this case we have a player who plays a  face man  a fictional persona who is
made up of other fictional personas. Yet even in this complicated situation of reced-
ing layers of fantasy, the role-player cannot create fictions of pure fiction. Instead,
Isaac draws on his knowledge both as a player and as a person to assist in the pro-
cess of creating the personas of his persona. As these examples illustrate, at a con-
ceptual level role-players may be able to draw fine distinctions between persona,
player, and person, but at a pragmatic level, these distinctions ultimately erode. In
the end, as Steve said,  I try to make everyone a little different. I don t want to play
a clone of the same character every time, though I do have personality traits that
creep in anyway.
CONCLUSION
Life, identity, and meaning are all understood as consisting of nothing more
than language games, exercises in role-playing. Social reality is experi-
enced through the performance of life, the performance of the everyday.
The only difference between the entertainment form known as the role-
playing game and the role-playing game of real life is that, for some reason,
a great deal of seriousness and levity is handed to each person in tandem
with the role they choose or are given.
 Daniel Mackay, The Fantasy Role-Playing Game
Participants in this study actively and playfully construct categorically unreal ficti-
tious personas in the process of playing a fantasy game. Fantasy role-playing games
also obligate participants to construct symbolic boundaries between person, player,
and persona. Yet, in practice, these conceptual distinctions failed; boundaries inevi-
tably implode as person, player, and persona blend and blur into an experience that
necessarily involves all three. In the end, even the most sophisticated role-players
found themselves in just the opposite situation players who play a multiplicity of
roles they cannot so easily compartmentalize. In this respect, fantasy role-playing
games are not unlike experiences of everyday life, nor are fantasy role-players necessar-
ily unique:  In taking on a role, the individual does not take on a personal, biographical
identity a part or a character but merely a bit of social categorization, that is, social
identity, and only through this a bit of his personal one (Goffman 1974:286).
352 Symbolic Interaction Volume 27, Number 3, 2004
Thus, on the one hand, role-playing games are whimsically distinct: it is reason-
able to assume that only a minority of people can claim experience at playing the
role of a dwarven barbarian. On the other hand, role-playing games are all too fa-
miliar: it is equally reasonable to assume that most people understand precisely
what it means to occasionally play other kinds of roles often of occupational or
institutional origin. In a similar sense, all is a ruse, charade, or game. In part, Goff-
man s insights originate from his acute awareness of how everyday life can be decon-
structed by these dynamics. Consistently locating people in the liminal threshold be-
tween illusion and reality, Goffman (1961b:84 152) has left a legacy that is fully
cognizant of the fact that  doing is not being. Individuals may  embrace a role and
 disappear completely into the virtual self available in the situation, to be fully seen in
terms of the image, and to confirm expressively one s acceptance of it (Goffman
1961b:106). Or, at another extreme, individuals may  distance themselves from a
role by creating  a wedge between the individual and his role, between doing and be-
ing (Goffman 1961b:108). Here, like elsewhere in his work, the hallmark of Goff-
man s modus operandi is to situate people in the margins between  playing and
 playing at (1961b:99). For this reason, although he may never have used these terms
in this way, he may have been the first to recognize and fully explore the precarious
distinctions between persona, player, and person in everyday life. Indeed, at the risk of
redundancy, at times we all find ourselves as participants in fantasy role-playing games.
Carse s (1986:177) brilliant analysis of society and culture through the lens of
finite and infinite games contains a fascinating parallel to Goffman. Both provide a
penetrating analysis of everyday life largely built of a single premise: roles are often
decidedly theatrical patterned, scripted, situated, and performed before an audi-
ence in accordance to social norms. Roles are also necessarily dramatic performed
by people who creatively play in a manner that persistently introduce elements of
indeterminacy and chance. Thus the structure of theater and the creative indetermi-
nacy of drama represent twin processes that, not by accident, mirror Mead s (1934)
classic distinction between the  I and the  me. One takes on a role theatrically
( me ), one enacts that role dramatically ( I ) neither necessarily subsumes the
other, nor is the whole of one s self found in one or the other. Instead, we, like par-
ticipants in fantasy role-playing games, find ourselves playing in the  cracks :
Without something to belong to, we have no stable self, and yet total commit-
ment and attachment to any social unit implies a kind of selflessness. Our sense
of being a person can come from being drawn into a wider social unit; our sense
of selfhood can arise through the little ways in which we resist the pull. Our status
is backed by the solid buildings of the world, while our sense of personal identity
resides in the cracks. (Goffman 1961a:320)
Carse (1986:177) concludes his analysis with a chapter that contains a single sen-
tence:  There is but one infinite game. This is the same conclusion that role-players
in this study have expressed. In spite of the heroic ways by which they distinguish
between fantasy and reality, persona and person, player and persona, person and
player, participants in role-playing games inevitably find themselves a part of  but
Role-Playing and Playing Roles 353
one infinite game. Finite boundaries and neatly crafted conceptual provinces of
meaning ultimately blend and blur to such an extent that nothing remains except a
player whose gaming activities include much more than the rolling of dice. In the
final analysis, it is doubtful that any of us can honestly claim otherwise. We all find
ourselves players located at the liminal margins between the people we believe we
are and the personas we play in various situated social encounters; between what
we believe we are and what we aspire to become; between what we believe we are
and what we believe others believe we are.
Have we gone too far with our implications? After all, the personas of role-playing
games belong, positively, to the realm of fantasy. Unlike everyday life, role-players
 adopt roles with which they strive to identify, but they do not fall victim to the illu-
sion that they are those roles (Mackay 2001:156; original emphasis). In spite of cer-
tain similarities, participants in role-playing games are also quite different from the-
ater actors: the role-player does not share bodies with the persona they play. While
they do share minds,  the player s body is never seen as the character s body
(Mackay 2001:88). Clearly, role-playing games neither represent nor imitate but
simulate, in which case perhaps we have not gone too far after all.
The role-playing game is a simulation that is not  of a territory, a referential being,
or a substance. It is the generation by models of a real without origin or reality: a
hyperreal (Baudrillard 1994:1). If the contemporary cultural landscape is awash
with the hyperreal, as some have suggested, then perhaps the fictions of role-playing
games represent something more than another example of the kind of blending and
blurring of fantasy, imagination, and reality that have long stimulated the interest of
symbolic interaction. In fantasy role-play, the blending and blurring of fantasy,
imagination, and reality is more than a conceptual, analytical, or methodological
strategy prevalent among theoretical perspectives with certain pragmatic roots; it is,
instead, an experiential condition endemic to a game with certain cultural roots. As
Holstein and Gubrium (2000:71) have suggested, in the hyperreal Disneyland of
contemporary culture, the self is  as much narratively constituted as actually lived ;
 self and its associated vocabulary are a living language game. Much like fantasy
role-playing games, in everyday life,  [w]ho we are ultimately taken to be as individ-
uals derives as much from the way we story ourselves, the textual material available
for storytelling, and the ways in which stories are  read and  heard, as from who
and what we might ostensibly be in our own rights. These, of course, are the inter-
textual contours of the self we live by (Holstein and Gubrium 2000:84, 205). In
fantasy role-playing games, participants literally construct a purely  narrative self ;
the fact that these games are fantasy does not obliterate the ways in which this pro-
cess is akin to the same dynamics in everyday life.
NOTES
1. Fantasy role-playing games are defined best by Mackay (2001:4 5; original emphasis):  [It is]
an episodic and participatory story-creation system that includes a set of quantified rules that
assist a group of players and a gamemaster in determining how their fictional characters spon-
354 Symbolic Interaction Volume 27, Number 3, 2004
taneous interactions are resolved. These performed interactions between the players and the
gamemaster s characters take place during individual sessions that, together, form episodes or
adventures in the lives of the fictional characters. . . . [T]he episodes become part of a single
grand story that I call the role-playing game narrative.
2. Dice rolls are used to simulate chance and probability. In this way role-playing games are sim-
ilar to games of alea one of four main classifications of games identified by Caillois ([1958]
2001). However, role-playing games are the antithesis of roulette or other true games of alea
in which  [t]he player is entirely passive; he does not deploy his resources, skill, muscles, or in-
telligence (Caillois [1958] 2001:17). Likewise, role-playing games, like card games, involve el-
ements of agon the use of  knowledge and reasoning that constitute the player s defense,
permitting him to play a better game (Caillois [1958] 2001:18). Yet neither alea nor agon is
adequate; role-playing games also involve significant mimicry, a  deploying [of] actions or sub-
mitting to one s fate in an imaginary milieu [and] becoming an illusionary character oneself,
and of so behaving (Caillois [1958] 2001:19). At best, one must concede that role-playing
games are a complex synthesis of these classic forms of play, if not something else altogether.
3. It is not quite accurate to call gamemasters  God-like ; in gaming situations that involve dei-
ties the gamemaster plays these gods as well. Gamemasters are above the gods: they create,
organize, and operate these fantasy worlds and mediate the supernatural forces that dictate
them. Within the frame of the game, it is not unfair to endow gamemasters with supreme status:
While players have control over their characters actions, the gamemaster has control over
the results of those actions. Life and death are in the gamemaster s hands. Furthermore, how
the characters perform in relation to the story (with its plot twists, villains, and so forth),
which most gamemasters script out before the session, will determine the rewards that the
gamemaster distributes to the characters. The pattern is the same one that Foucault ob-
served[;] . . . the gamemaster is always present in a panoptic position, which inscribes itself
onto the performing consciousness of the player. The gamemaster, in fact, is in an even more
enviable position than the guard who watches the prisoners from the tower in utilitarian phi-
losopher Jeremy Bentham s 1787 blueprints for the Panopticon. . . . The game master is usu-
ally in a privileged position of observation in relation to the players. . . . This receptive ca-
pacity allows the gamemaster to survey each character, the group of characters, and the
players. . . . Without a doubt the gamemaster holds the most power. (Mackay 2001:94 97).
4. The distinction between person, player, and persona adheres to what Fine (1983:194, 205) de-
scribes as the  three basic frames that operate in fantasy gaming. As Fine wrote, each of these
frames  has a world of knowledge associated with it the world of commonsense knowledge
grounded in one s primary framework, the world of game rules grounded in the game struc-
ture, and the knowledge of the fantasy world. Fine s investigation is solidly supported by his
use and extension of Goffman s (1974) Frame Analysis. Goffman (1974:129) also clearly dif-
ferentiates person, player, and persona:  The difference between actual and scripted becomes
confused with the difference between personal identity and specialized function, or (on stage)
the difference between part and capacity. I shall use the term  role as an equivalent to special-
ized capacity or function, understanding this to occur both in offstage, real life and in its
staged version; the term  person will refer to the subject of a biography, the term  part or
 character to a staged version thereof.
5. Others have also noted how Mead s framework necessarily entails richly layered, interrelated,
and thoroughly inseparable elements of fantasy, imagination, and reality. Stone (1970) pro-
vides one of the clearest articulations of these relationships in his discussion of  fantastic so-
cialization. He identifies two kinds of socialization that can be found in Mead s  play. The
first is widely noted by sociologists: genuine  anticipatory socialization. Here realistic roles
are acted according to expectations that one would reasonably expect to be adopted or en-
countered later in life. The second,  fantastic socialization, is often overlooked: here, one
entertains roles that can seldom if ever be expected or adopted. Stone provides the example
of children playing cowboy or Indian. We may add to this a long list of superheroes, dead his-
torical figures, media-produced characters, and others who clearly occupy a central role in the
 fantastic socialization of all of us.
Role-Playing and Playing Roles 355
6. Other contemporary interactionist studies have also emphasized these dynamics. Recent studies
of cybersex (Waskul 2002; Waskul, Douglass, and Edgley 2000) have explored how partici-
pants evoke an often fantastical  virtual body for the paradoxical purposes of having sex in
a disembodied communication environment. Although sometimes amusingly fictional, these
virtual bodies still function as a kind of discursive looking-glass in which selfhood is reflected.
In a recent study of Russian Jewish immigrants, Rapoport, Lomsky-Feder, and Heider (2002)
document how immigrants recollections of past experiences of anti-Semitism inform their
process of entering a new society and identity construction. Rapoport et al. illustrate how rec-
ollections are fashioned into a  memory kit (p. 176) where  the present draws on the past
selectively and the past is not literally constructed (p. 180). Immigrants actively construct
these memory kits to such an extent that the  past loses its concreteness and instead becomes
a reservoir of private and collective memories containing an array of multipurpose resources
for interpretation:  the memory kit that the immigrants carry with them consists of versatile
ready-to-use narratives that render them free to maneuver between different interpretations
of anti-Semitism (p. 182).
7. In Dungeons & Dragons there are three core rulebooks (and a massive supply of other publi-
cations) that specify guidelines for character classes, fantasy humanoid races, medieval weap-
ons and armor, magical spells, skills, abilities, movement, mythical monsters, and supernatural
forces that include powerful competing gods (to list a few major categories). From these
guidelines players craft characters within shared fantasies (Fine 1983): a vast cosmos of collec-
tively constructed imaginary actions, interactions, reactions, and the myriad consequences that
result from fantasy events.
8. We use male pronouns in this study because all the participants in our study are men. Fine
(1983) discusses this gender bias in fantasy role-playing games a bias that, in our completely
unrepresentative and very localized sample, appears to remain unchanged in the two decades
since Fine s ethnography was originally published. We admit the possibility that gender has
influenced the dynamics explored in our research.
9. As Goffman (1974: 23) notes,  With the possible exception of pure fantasy or thought, what-
ever an agent seeks to do will be continuously conditioned by natural constraints, and that ef-
fective doing will require the exploitation, not the neglect, of this condition. . . . [T]he assump-
tion is, then, that although natural events occur without intelligent intervention, intelligent
doings cannot be accomplished effectively without entrance into the natural order. From this
perspective, role-playing games also become  life-like because the actions of player-characters
are always subject to the outcome of random dice rolls that mock these  natural constraints.
10. All participants knew we were conducting a study of fantasy role-playing games. However,
some participants apparently disregarded the fact that the first author is a professor at the
local university and, since he does not look much different from a student, the occupational
role seemed easy to forget or ignore. We did not conceal our intentions or identity, but without
formal introductions the situation was sometimes ambiguous. During gaming sessions, we were
players; other players did not seem to care about these kinds of insignificant and distracting
details. The full extent of the irrelevance of these occupational roles was illustrated after over
a of month game-play. The gamemaster made a casual remark about the first author being a
professor at the university. Somewhat surprised, a younger player asked what the first author
teaches. When he responded,  Sociology, the player merely said,  Cool. I think I ll take that
next year when I m at the university. The subject of our occupation never came up before or
again. It simply did not matter.
REFERENCES
Baudrillard, Jean. 1983.  What Are You Doing after the Orgy? Traverse, 29 October, pp. 1 7.
Caillois, Roger. [1958] 2001. Man, Play and Games. Chicago: University of Chicago Press.
Carse, James. 1986. Finite and Infinite Games. New York: Macmillan.
356 Symbolic Interaction Volume 27, Number 3, 2004
Cohen, Stanley and Laurie Taylor. 1992. Escape Attempts: The Theory and Practice of Resistance to
Everyday Life. 2d ed. New York: Routledge.
Cook, Monte, Jonathan Tweet, and Skip Williams. 2000. Dungeons & Dragons Player s Handbook:
Core Rulebook I. Renton, WA: Wizards of the Coast, Inc.
Cooley, Charles Horton. [1902] 1964. Human Nature and Social Order. New York: Scribner s.
Fine, Gary Allen. 1983. Shared Fantasy: Role-Playing Games as Social Worlds. Chicago: University
of Chicago Press.
Goffman, Erving. 1961a. Asylums: Essays on the Social Situation of Mental Patients and Other
Inmates. Garden City, NY: Doubleday Anchor.
   . 1961b. Encounters: Two Studies in the Sociology of Interaction. Indianapolis, IN: Bobbs-
Merrill.
   . 1974. Frame Analysis. Cambridge, MA: Harvard University Press.
Hertz, Rosanna. 2002.  The Father as an Idea: A Challenge to Kinship Boundaries by Single
Mothers. Symbolic Interaction 25(1):1 31.
Holstein, James and Jaber Gubrium. 2000. The Self We Live By: Narrative Identity in a Postmodern
World. Cambridge, MA: Oxford University Press.
Huizinga, Johan. 1950. Homo Ludens: A Study of the Play Element in Culture. Boston: Beacon
Press.
Mackay, Daniel. 2001. The Fantasy Role-Playing Game: A New Performing Art. Jefferson, NC:
McFarland.
Maines, David, Noreen Sugrue, and Michael Katovich. 1983.  The Sociological Import of G. H.
Mead s Theory of the Past. American Sociological Review 48(2):161 73.
Mead, George Herbert. 1934. Mind, Self, and Society. Edited by C. Morris. Chicago: University of
Chicago Press.
   . 1936. Movements of Thought in the Nineteenth Century. Chicago: University of Chicago
Press.
Rapoport, Tamar, Edna Lomsky-Feder, and Angelika Heider. 2002.  Recollection and Relocation
in Immigration: Russian-Jewish Immigrants  Normalize Their Anti-Semitic Experiences.
Symbolic Interaction 25(2):175 98.
Schutz, Alfred. [1932] 1967. The Phenomenology of the Social World. Evanston, IL: Northwestern
University Press.
   . 1973. Collected Papers I: The Problem of Social Reality. The Hague: Martinus Nijhoff.
Simmel, Georg. [1911] 1971.  The Adventurer. Pp. 187 98 in Georg Simmel: On Individuality and
Social Forms, edited by D. Levine. Chicago: University of Chicago Press.
Smith, Geoffrey. 1980.  Dungeons and Dollars. Fortune, September 15, pp. 137 42.
Stone, Gregory. 1970.  Appearance and the Self. Pp. 394 414 in Social Psychology through Sym-
bolic Interaction, edited by G. Stone and H. Farberman. Waltham, MA: Xerox College Pub-
lishing.
Turner, Ralph. 1978.  The Role and the Person. American Journal of Sociology 84:1 23.
Waskul, Dennis. 2002.  Naked Self: Being a Body in Televideo Cybersex. Symbolic Interaction
25(2):199 227.
Waskul, Dennis, Mark Douglass, and Charles Edgley. 2000.  Cybersex: Outercourse and the
Enselfment of the Body. Symbolic Interaction 23(4):375 97.
Zurcher, Louis. 1983. Social Roles: Conformity Conflict and Creativity. Beverly Hills, CA: Sage.